The goal of this Project is to develop personalized therapeutic approaches for NSCLC patients based on tumor molecular profiles taken prior to treatment. Currently, molecularly-guided therapy for NSCLC patients is limited to the minority of patients with targetable oncogenic drivers (e.g. EGFR mutations) but the majority of patients to not have such alterations with matching drugs, and these markers to not inform the selection of chemotherapy or other treatments. This project will create a broadly useful classification for NSCLC patients using two approaches. Expression Clades (ECs), based on mRNA expression profiles, and Mutational Clades (MCs), based on DNA mutational alterations. We hypothesize that: 1) tumor ECs and MCs alone, or when combined together (Genomic Clades, GCs), reveal underlying biologically distinct lung cancer subgroups with different therapeutic responses to targeted agents, chemotherapy and acquired vulnerabilities (synthetic lethalities); 2) clades may be used to predict treatment response and facilitate developing novel targeting strategies. Our preliminary data support these hypotheses as well as the feasibility of applying the clades to the clinic. The translational goals are to develop these clades as CLIA-certified enrollment biomarkers for such personalization and to study them in preclinical models, patient tumor specimens, and in a window of opportunity neoadjuvant trial. We have developed the following specific aims to bring this project to fruition. Aim 1: We will develop and refine our classification of ECs, MCs, and GCs using clinical and molecularly annotated data sets (e.g. TCGA Lung datasets, which we have profiled for a panel of protein markers), and apply the clades to our existing preclinical models including cell lines and xenograft models. Aim 2. We will test and validate the association between clades and drug response for selected targeted agents (nintedanib, sorafenib) and chemotherapy regimens in preclincial in vitro and in vivo models, and identify novel clade-based targeting strategies and molecular vulnerabilities. Using this approach we have recently identified novel monogenic vulnerabilities for specific clades. Aim 3. We will translate clades into the clinic by testing their value in predicting prognosis and benefit from adjuvant, chemotherapy, as well as treatment response to targeted agents (nintedanib and sorafenib) and nintedanib in combination with chemotherapy in the neoadjuvant trial. To do this we will leverage our existing SPORE Pathology Core resources, those from the completed BATTLE study, and those from this project's prospective window of opportunity neoadjuvant therapy trial. We have assembled a multidisciplinary team of laboratory and clinical investigators including collaborations with all ofthe SPORE Cores.